The Development of the Research in Digital Visibility Instruments and the Related
Comparative Experiments
J.L. Wang
a,*
, W.F. Zhao b, J.C. Liu c, G.H. Liu b, X.Y. Cao a, S.X. Ruan a
a
Institute of Urban Meteorology, CMA, Beijing, China, 100089 – (jlwang, xycao, sxruan)@bjmb.gov.cn
b
Beijing Meteorological Information Center, Beijing, China, 100089, – (wfzhao, ghliu)@bjmb.gov.cn
c
Technological Equipment Center of Beijing Meteorological Bureau, Beijing, China, 100089, – jcliu@bjmb.gov.c
Abstract—The observation automation of cloud, visibility,
and weather phenomena are principal issue to solve in the
modern meteorological observation field. In the late 20th
century, scientists have started to develop a kind of digital
visibility automatic monitor, which has adopted advanced
digital photographic technology and image recognition
technology, based on the definition of visibility and fully
imitating the human eye observing visibility principle
Steffens C 1949 . Undoubtedly, the digital visibility
automatic monitor is the best instrument to replace the
manual observation of visibility
homas 1994 . This
article briefly introduces the methods of visual and
instrumental observation of visibility, and elaborates the
basis of the Digital Photograph Device System
Xingsheng 1999 , its latest development, and the
comparative results between Finnish Vaisala visibility
instrument and manual observation Jingli 2002 . In
the end, the article reaches an error analysis conclusion and
discusses about the problems and further improving
measures.
（
，
）
（
（
，
）
）
（
， ）
Keywords: Digital Photograph Visibility Device System
(DPVS); Visibility; Digital Photography; Comparative
Experiment.
1.
Considering the visibility observation status at home and abroad
mentioned above, it is imminent to develop an automatic
visibility observational instrument to fully replace the manual
observation. In the late 20th century, in China, scientists have
started to develop a kind of Digital Photograph Device
System(DPVS), that is, digital visibility instrument, which has
adopted advanced digital photographic technology, fully imitates
the human eye observing visibility principle, and is designed
based on the definition of visibility. Undoubtedly, digital
visibility instrument is the best instrument to replace the manual
（
，
）
Jingli 2006 . With the rapid
observation of visibility
development of digital photography technology, especially that
the pixel of the CCD camera has earlier reached the resolution of
the human eye (10 million pixels), while the price is in freefall, it
is certainly a general trend to use digital photography technology
to measure atmospheric visibility.
2.
INSTRUMENTAL VISIBILITY OBSERVATION
Definition of Visibility that is the maximum distance from which
in a weather condition the human eye(standard visual acuity) can
see and identify a target’s contour (black, with appropriate size,
and a general view of 0.5 to 5 °) in the background of sky. Figure
1 shows that Visibility Differences
.
INTRODUCTION
At present, China's meteorological station still adopts the
traditional manual observation method to observe the basic
meteorological elements such as clouds, visibility, and weather
phenomena. This kind of method is very subjective and noncomplete, thus it is hard to apply it to quantitative / model
system. In the instrumental observation of visibility, the main
equipments adopted include transmission visibility meter,
automatic laser visibility observational instrument, and forward
scattering visibility instrument. Atmospheric transmission
instrument is used to measure attenuation coefficient, that is, to
let a light beam run through the atmospheric column between
two fixed points to measure the transmittance of the
atmospheric column, thus to assess its visibility value. In this
sort of measurement, the light beam is supposed to run through
an atmospheric column that is sufficiently long. However, in
rainy, foggy and other low-visibility days, there will be a greater
error due to water vapor absorption and other complex
conditions, thus it has very big limitation1.. While automatic
laser visibility observational instrument is to extrapolate
visibility from the atmospheric light extinction coefficient
measured by laser, which is relatively more objective and
accurate. But this method is still limited to a small number of
research departments, because the laser radar is very expensive,
and its maintenance cost is high too; the operation of the
equipment is complicated, and in rainy and foggy days, it is also
difficult to carry out normal observation.
Figure 1. Visibility Differences.
2.1 Transmission visibility meter
Transmission visibility meter is to measure directly the
transmittance of the atmospheric column between two fixed
points when a collimated light beam runs through, and estimate
the average extinction coefficient from the transmittance, thus
obtain the visibility value, that is, the meteorological optical
visual range. It is obviously accurate to infer the visibility within
a distance of about several kilometers or dozens of kilometers
from the transmittance of the atmospheric column that has only a
limited base length of dozens of meters. (In this method, the light
beam is supposed to run through an atmospheric column that is
sufficiently long). And in the rainy, foggy and other lowvisibility days, there will be a greater error due to water vapor
absorption and other complex conditions, thus it has very big
limitation. Figure 2 shows that schematic for Transmission
visibility meter.
absorption and other complicated conditions. Figure 4 shows
that schematic for Scattering visibility instrument
Figure 2. Schematic for Transmission visibility meter
2.2 Laser Visibility Instrument
Automatic laser visibility observational instrument is to
measure the echo energy of the laser in the process of
transmitting in the atmosphere. Therefore the extinction
coefficient can be calculated due to the fact that the extinction
coefficient is in proportion to the echo energy. In turn, the
visibility can be inferred from the extinction coefficient. The
atmospheric extinction coefficient obtained here is based on the
premise that the air on the optical transmission path is evenly
distributed, and the molecular and aerosol absorption is
negligible. As the baseline is longer, the observation is
relatively more objective. However, as the absorption is
neglected, in rainy and foggy days, the accuracy of the
observation will certainly be affected. The laser radar is very
expensive, and its maintenance cost is high too; the operation of
the equipment is complicated; therefore this method is still
limited to a small number of research departments, and in rainy
and foggy days, it is also difficult to carry out normal
observation. Figure 3 shows that schematic for Laser Visibility
Instrument.
Figure 4. Schematic for Laser Visibility Instrument
2.4 Digital Visibility Meterr
Digital visibility meter is a new instrument which adopts digital
photographic method to measure visibility. The design idea is
proposed by Mr. Zhou Xiuji, academician of Chinese Academy
of Sciences, renowned atmospheric physicist. It is entirely
developed from the principles of manual visibility observation.
According to manual visibility observation method, first, choose
several targets which are clear and can be identified by CCD in
normal weather conditions, and then use CCD to take photos of
the selected targets to get the images which are then conveyed
to computer through image acquisition card. Computer will then
analyze and deal with the image data of the targets respectively,
substituting the image data into the relevant formulas to work
out the observational visibility value according to the principles
of manual visibility observation and the definition of visibility.
Figure 5 shows that schematic for Digital Visibility Meter.
Figure 5. Schematic for Digital Visibility Meter
Figure 3. Schematic for Laser Visibility Instrument
2.3 Scattering visibility instrument
Scattering visibility instrument is to measure the scattered light
intensity of a limited volume (small volume) of air, and then
estimate the atmospheric light extinction coefficient to figure
out the atmospheric visibility. The calculation of extinction
coefficient is based on the following three assumptions: the
atmosphere is homogeneous, that is, the atmosphere is
uniformly distributed; atmospheric light extinction coefficient
R is equal to the scattering of fog, haze, snow and rain in the
atmosphere, that is to say, molecular absorption, scattering, or
the interactive optical effect within the molecule is zero; the
scattered light intensity measured by the scattering instrument is
in direct proportion to the scattering coefficient. Under normal
circumstances, select an appropriate angle, scattering signal is
approximately in direct proportion to the scattering
coefficient[2]. Apparently all the three assumptions mentioned
above are untenable in the actual observation environment.
Uniform air distribution is impossible, and it is unreliable to
extrapolate the atmospheric visibility of a distance around a few
or even dozens of kilometers from the extinction coefficient of a
limited volume of air. Atmospheric light absorption can not be
ignored, especially in rainy, foggy and other low-visibility
weather; a bigger error will be caused by the water vapor
3. ASSESSMENT OF INSTRUMENTAL
OBSERVATIONAL VISIBILITY VALUE AND ERROR
ANALYSIS
As noted above, the former three devices are to estimate the
visibility value by speculating atmospheric extinction coefficient
indirectly. Transmission visibility meter is to measure directly
the transmittance of the atmospheric column and estimate the
extinction coefficient from the transmittance, thus obtain the
visibility value.
Automatic laser visibility observational
instrument is generally to measure the latter-part scattering
coefficient to infer the extinction coefficient and in turn obtain
the visibility value. Scattering visibility meter is usually to
measure the scattering light intensity of a limited volume of air
directly and thus estimate the atmospheric extinction coefficient
to figure out the visibility value. Digital visibility meter is based
on the principles of manual visibility observation and the
definition of visibility, and it is to figure out the visibility value
from the target distance and clarity, which is therefore the best
instrument to replace manual visibility observation.
4. COMPARATIVE EXPERIMENTS
In 2008, Institute of Urban Meteorology, Beijing, China
Meteorological Administration, has transformed and laid five
sets of automatic digital photographic visibility monitors. from
January 1, 2009 to March 17, 2009, comparative experiments
between DPVS and Finnish Vaisala visibility meter as well as
manual visibility observation have been carried out for 75 days
during daytime simultaneously in five stations. In Miyun station,
comparative experiments between DPVS and Vaisala FD12
visibility meter are carried out. The results of the comparative
experiments show that the relative standard deviation between
DPVS and Vaisala FD12 visibility meter is 20% or close to 20%.
Table 1 shows the comparative result of the relative standard
deviation between the visibility observed by DPVS and FD12.
Formula (1) is the relative standard deviation calculation
formula:
N
∑ [( X
i
− Z i ) / Z i ]2
i =1
σ rd =
× 100%
N
（1 ）
In formula (1) , Xi represents the observational visibility value of
，
DPVS Zi represents the standard FD12 observational visibility
value, and N represents the times of observations.
σ rd
represents the relative standard deviation between DPVS and
FD12.
THE COMPARATIVE RESULT
TABLE I.
Visibility Scope(km)
Comparative
Result
Below 2
2-5
5-10
10-20
Above 20
number
116
692
1628
1222
2262
11.5
17.8
21.7
23.1
2.1
σ rd
a
(%)
a. Relative Standard Deviation
ACKNOWLEDGMENT
This research work were mainly supported by The Ministry of
Science and Technology of the People’s Republic of China (JG2003-21),China
Meteorological
Administration
(CMATG2006M13), the public (Meteorological sector) research
special funds (GYHY200806027), Ministry of Finance People’s
Republic of China (401), and Urban meteorological research
Foundation (No. UMRF200906). The authors would like to
thank the participating personnel, as well as expressing our
appreciation to anonymous reviewers for constructive comments
on the manuscript.
REFERENCES
Steffens C. “Measurement of visibility by photographic
photometry,” Industrial Engi-neering Chemistry, vol. 41, pp.
2396-2399, January 1949.
Thomas Legal, Louis Legal and Waldemar Lehn. “Measuring
Visibilty Using Digital Remote Video Cameras,” American
Meteorological Society, 9th symp on met observ & instr, pp.
87–89, 1994.
Xingsheng Xie, Shanchang Tao and Xiuji Zhou. “Observation
of Meteorological Visibility by Digital Photographic Method,”
Chinese Science Bulletin, vol. 44, pp. 97-100, January 1999 (In
Chinese) .
Jingli Wang, Conglan Cheng and Xiaofeng Xu. “Comparative
Experiments on Digital Photograph Device System.
Meteorological Science and Technology,” vol. 30, pp. .253-257,
November 2002 (In Chinese) .
Jingli Wang, Xulin Liu and Wufeng Gao, “The Design for
Control Circuits of a Digital Photography Visiometer System,”
Meteorological Science and Technology, vol. 34, pp. 633-637,
September 2006 (In Chinese) .